The samples this guy is interested in are synthetic. No Cu visible. Cu XANES tends not to have huge white lines, which is what would be required for the contamination to show as a peak with no detectable edge jump. Also, the peak is nowhere near the Cu K-edge. For similar reasons, it's not Fe either. mam On 8/4/2016 4:20 PM, Matt Newville wrote:

Hi Matthew,

On Thu, Aug 4, 2016 at 2:24 PM, Matthew Marcus mailto:mamarcus@lbl.gov> wrote:

Does anyone know of data on multi-electron peaks at the Yb L3 edge? One of the users here was just running Yb and finds on every spectrum a bump at the same energy. It's not a Bragg glitch from the sample or an I0 glitch or any other obvious artifact. It reminds me of the two little bumps at Ce. If someone has analyzed and parameterized this peak as a fraction of edge jump, then it would be possible to subtract it off. The attached papers show exmples for almost every lanthanide *execpt* Yb.

I have not done much work on Yb, and all the measurements I can find are on the L2 edge, not the L3 edge.

That's because the L3 edge is at 8944, 30 to 40 eV below the Cu K edge. In my experience, it's really hard to get rid of Cu and Fe as trace contaminants, especially in natural samples. Is it possible that what you're seeing is not multi-electron excitation by Cu contamination?

For those doing Ce, I've written a program which does the correction, essentially as presented in the attached Gomilsek paper.

That sounds useful.

--Matt

_______________________________________________ Ifeffit mailing list Ifeffit@millenia.cars.aps.anl.gov http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit Unsubscribe: http://millenia.cars.aps.anl.gov/mailman/options/ifeffit

It doesn't. See the attached, which is a slide I put in that user's RunNotes file. I set up ppt files for each visit by each user but not all of them bother to record their stuff there. Anyway, you can see the peak I'm referring to, which just about matches Dy L1. The other DyL edges are below the start of data. I'm not all that surprised to find a little of another lanthanide; not only is it hard to separate them, but we've seen incidents in that lab of cross-contamination. For instance, one of his samples had more Lu in it than the intended Yb, mam On 8/4/2016 8:01 PM, Matt Newville wrote:

Hi Matthew,

Attached is a file of Yb L3 edge on something that was labeled "Yb AC" and measured in fluorescence many years ago, almost certainly in solution. I suspect that "AC" means "acetate".

This is a plain ASCII file (not XDI!). The data quality is not great. Does this have the feature you're seeing?

--Matt

_______________________________________________ Ifeffit mailing list Ifeffit@millenia.cars.aps.anl.gov http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit Unsubscribe: http://millenia.cars.aps.anl.gov/mailman/options/ifeffit

Hi Matthew, No one ever expects lutetium. --Matt